In 2003, evolutionary biologist Richard Lenski, philosopher Robert Pennock and others co-published a Nature paper titled “The evolutionary origin of complex features” reporting results of a computer simulation of evolution dubbed “Avida.” Though publicly arguing that Avida refuted intelligent design by showing the evolution of irreducible complexity, their paper refused cite the work of Michael Behe or any other ID proponent. Now, Winston Ewert, William Dembski, and Robert Marks expose in a paper in Proceedings of the 2009 IEEE International Conference on Systems, Man, and Cybernetics why Lenski and Pennock’s “Avida” simulation fails to accurately model Darwinian evolution.
Darwinian evolution has no prior knowledge about the search target, but Avida’s programmers have intelligently designed Avida by smuggling in “active information” to help the program overcome the handicap of Darwinian blindness. Avida is based upon the premise that its target function (“EQU”) will be eventually found simply by building on simpler logic functions. Ewert, Dembski, and Marks call this attempt to model a stepwise advantage “stair step active information,” observing that “Avida uses stair step active information by rewarding logic functions using a smaller number of nands to construct functions requiring more.” Significantly, Ewert, Dembski, and Marks find that “Removing stair steps deteriorates Avida’s performance,” quoting from Lenski and Pennock’s paper admitting that “where only EQU was rewarded … none of these populations evolved EQU.” Avida is thus designed to evolve, even though its designers don’t make that clear. Ewert, Dembski, and Marks thus conclude with the exhortation that, “To have integrity, computer simulations of evolutionary search like Avida should make explicit … the prior knowledge that gives rise to the active information in the search algorithm.”
The abstract reads:
According to conservation of information theorems, performance of an arbitrarily chosen search, on average, does no better than blind search. Domain expertise and prior knowledge about search space structure or target location is therefore essential in crafting the search algorithm. The effectiveness of a given algorithm can be measured by the active information introduced to the search. We illustrate this by identifying sources of active information in Avida, a software program designed to search for logic functions using nand gates. Avida uses stair step active information by rewarding logic functions using a smaller number of nands to construct functions requiring more. Removing stair steps deteriorates Avida’s performance while removing deleterious instructions improves it. Some search algorithms use prior knowledge better than others. For the Avida digital organism, a simple evolutionary strategy generates the Avida target in far fewer instructions using only the prior knowledge available to Avida.
(Winston Ewert, William A. Dembski, and Robert J. Marks II, “Evolutionary Synthesis of Nand Logic: Dissecting a Digital Organism,” Proceedings of the 2009 IEEE International Conference on Systems, Man, and Cybernetics San Antonio, TX, USA, 3047-3053 (October 2009).)